当前位置:诺佳网 > AI人工智能 > AI通用技术 >

AI生成视频检测方法及其相关研究

时间:2025-05-03 | 栏目:AI通用技术 | 点击:

【1】AI-Generated Video Detection via Spatio-Temporal Anomaly Learning

AI生成视频和真实视频的光流图
AIGVDet
Robustness evaluation

【2】DeCoF: Generated Video Detection via Frame Consistency

生成视频中空间伪影与时间伪影

【2.1】Spatiotemporal Convolutional Neural Networks (STCNN) rely on spatial artifacts

【2.2】Capturing Universal Spatial Artifacts is Difficult

t-SNE可视化图,四种视频生成模型相关的真实视频帧与伪造视频帧的分布情况
真实视频帧与由不同视频生成模型生成的伪造视频帧的平均频谱对比图
DeCoF

【3】DeMamba: AI-Generated Video Detection on Million-Scale Benchmark

GenVideo数据集中真实视频和生成视频的统计数据
Detail Mamba

【4】GenVidBench: A Challenging Benchmark for Detecting AI-Generated Video

在这里插入图片描述

【5】Distinguish Any Fake Videos: Unleashing the Power of Large-scale Data and Motion Features

在这里插入图片描述

【6】On Learning Multi-Modal Forgery Representation for Diffusion Generated Video Detection

【7】What Matters in Detecting AI-Generated Videos like Sora?

【8】Turns Out I’m Not Real: Towards Robust Detection of AI-Generated Videos

在这里插入图片描述

【9】Beyond Deepfake Images: Detecting AI-Generated Videos

【10】Exposing AI-generated Videos: A Benchmark Dataset and a Local-and-Global Temporal Defect Based Detection Method

Acknowledgments

您可能感兴趣的文章:

相关文章